Smart soldiers can adapt quickly to changes in unfamiliar and/or dangerous terrain, but they may find it easier when their means of transportation can help with the thinking. That is the thought behind the U.S. Army Combat Capabilities Development Command’s (CCDC) Army Research Laboratory Artificial Intelligence for Maneuver and Mobility (AIMM) program. By equipping autonomous vehicles with suitable AI capabilities, the vehicles can serve as robotic combat partners, increasing the range and power of the ground troops.
AIMM program manager Dr. John Fossaceca hopes to provide enough AI within the next generation of autonomous combat vehicles to enable it to operate independently of the main combat vehicle on any mission: “The main purpose of this essential research program is to build autonomous systems that help the Army effectively execute multi-domain operations.” He expects some amount of human interaction with the vehicles, but added: “We don’t want soldiers to be operating these remote-controlled vehicles with their heads down, constantly paying attention to the vehicle to control it. You want these systems to be fully autonomous so that these soldiers can do their jobs and these autonomous systems can work as teammates and perform effectively in the battlefield.”
The expected operating environment is one of the factors that helps differentiate autonomous vehicles developed for the Army versus those for commercial customers. Commercial self-driving vehicles are expected to handle well-maintained public roads and highways while autonomous vehicles for Army use will often face rough and unpaved roads and severe operating conditions, such as covering significant distances through a desert. Next-generation autonomous vehicles for the Army are expected to have sufficient AI to not only perform self-driving functions but to work with soldiers on their own decision making, almost like in-field computers on wheels.
Fossaceca elaborated: “Future military missions are going to require autonomous vehicles that can determine what the passable routes might be, calculate the best route and make assessment about what’s happening in the environment. We want to integrate this autonomous behavior in modern military vehicles so that it feels less alien to the soldier and can decrease the cognitive burden of the decision-making process.” To accomplish these goals, researchers in the program have put considerable effort into improving the autonomy software stack, a collection of algorithms, libraries, and software components supporting advanced behavior for autonomous vehicles. The algorithms will equip an autonomous vehicle with a “virtual brain,” enabling a host of different functions including navigation, control, planning, perception, and reasoning so that the vehicle can respond to changing conditions.